[journal of Asian Culture] Controll and Null Anaphora in Japanese

One of the most controversial problems in recent work on syntax is the so-called control phenomenon of the subject of an infinitival clause; More specifically, the central issue is how the subject of the infinitival clause ia interpreted. For instance, in the sentence, 'John promised Mary to go to Japan,' the subject of the infinitival clause, 'to go to Japan,' is necessarily interpreted as the same aa the matrix subject, 'John'; whereas, in 'John persuaded Mary to go to Japan,' the subject of the infinitival clause is always construed as the same as object of the matrix clause, 'Mary.' Various linguists have attempted to determine which component of grammar this obligatory control phenomenon belongs to. Different analyses proposed to account for thia phenomenon in English include the syntactic/pragmatic approach of Manzini (1983), the syntactic/thematic/pragmatic analysis of Nishigauchi (1984) , and at least one approach, that of Culicover & Wilkins (forthcoming) which uses thematic relations. In this paper, I will discuss obligatory control phenomenon in Japanese, and examine what kind of analysis will be plausible. Consequently, this paper should be viewed as programmatic, rather than definitive. The range of data I will be discussing is in (1-7) •1 (1) is an example of 'subject' control. roNTROL & NULL ANAPHORA 53